Add Tium's five models to model-metadata.json - #5697
Open
curiousbox wants to merge 1 commit into
Open
Conversation
Tium is an OpenAI-compatible gateway for open-weight models. Reached through OPENAI_API_BASE, its models are named without a provider prefix, for example `glm-5.3-flash`. That matches no entry in litellm's cost map, so aider reports "Unknown context window size and costs" and a session cost of $0.00. The `openai/` keys are what that path looks up, which is why a cost-map entry under a vendor prefix does not cover it. litellm already catalogues some of these models under `zai/` and `openrouter/z-ai/`; those resolve for users of those providers, not this one. Every figure is derived from the gateway's GET /v1/models rather than typed in: input is the published rate times the model multiplier, output is input times the published output weight. Context and max output are the served ceilings, not the models' upstream maximums, which are higher. The two DeepSeek entries are marked supports_tool_choice: false because DeepSeek refuses tool_choice="required" while reasoning is enabled. Function calling and prompt caching work on all five. Verified by running aider against the modified file with no local .aider.model.metadata.json present: the warning is replaced by a resolved model.
|
|
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Adds five Tium models to
model-metadata.json.Tium is an OpenAI-compatible gateway for open-weight models. Reached through
OPENAI_API_BASE, its models are named without a provider prefix, for exampleglm-5.3-flash. That matches no entry in litellm's cost map, so aider reports "Unknown context window size and costs" and a session cost of $0.00.The
openai/keys are what that path looks up, which is why a cost-map entry under a vendor prefix does not cover it. litellm already catalogues some of these models underzai/andopenrouter/z-ai/; those resolve for users of those providers, not this one.Every figure is derived from the gateway's
GET /v1/modelsrather than typed in: input is the published rate times the model multiplier, output is input times the published output weight. Context and max output are the served ceilings, not the models' upstream maximums, which are higher.The two DeepSeek entries are marked
supports_tool_choice: falsebecause DeepSeek refusestool_choice="required"while reasoning is enabled. Function calling and prompt caching work on all five.Verified by running aider against the modified file with no local
.aider.model.metadata.jsonpresent: the warning is replaced by a resolved model.